Over the past few years, the world’s major breweries and glass packaging users have been demanding significant reductions in the carbon footprint of packaging materials, following the megatrend of reducing plastic use and reducing environmental pollution. For a long time, the task of forming the hot end was to deliver as many bottles as possible to the annealing furnace, without much concern for the quality of the product, which was mainly the concern of the cold end. Like two different worlds, the hot and cold ends are completely separated by the annealing furnace as the dividing line. Therefore, in the case of quality problems, there is hardly any timely and effective communication or feedback from the cold end to the hot end; or there is communication or feedback, but the effectiveness of the communication is not high due to the delay of the annealing furnace time. Therefore, in order to ensure that high-quality products are fed into the filling machine, in the cold-end area or the quality control of the warehouse, the trays that are returned by the user or need to be returned will be found.
Therefore, it is particularly important to solve product quality problems in time at the hot end, help molding equipment increase machine speed, achieve lightweight glass bottles, and reduce carbon emissions.
In order to help the glass industry achieve this goal, XPAR company from the Netherlands has been working on developing more and more sensors and systems, which are applied to the hot-end forming of glass bottles and cans, because the information transmitted by the sensors is consistent and efficient. Higher than manual delivery!
There are too many interfering factors in the molding process that are affecting the glass manufacturing process, such as cullet quality, viscosity, temperature, glass uniformity, ambient temperature, aging and wear of coating materials, and even oiling, production changes, stop/start The design of the unit or the bottle can affect the process. Logically, every glass manufacturer seeks to integrate these unpredictable disturbances, such as gob state (weight, temperature and shape), gob loading (speed, length and time position of arrival), temperature (green, mold, etc.) , punch/core, die) to minimize the impact on molding, thereby improving the quality of glass bottles.
Accurate and timely knowledge of gob status, gob loading, temperature and bottle quality data is the fundamental basis for producing lighter, stronger, defect-free bottles and cans at higher machine speeds. Starting from the real-time information received by the sensor, the real production data is used to objectively analyze whether there will be later bottle and can defects, instead of various subjective judgments of people.
This article will focus on how the use of hot-end sensors can help produce lighter, stronger glass jars and jars with lower defect rates, while increasing machine speed.
This article will focus on how the use of hot-end sensors can help produce lighter, stronger glass jars with lower defect rates, while increasing machine speed.
1. Hot end inspection and process monitoring
With the hot-end sensor for bottle and can inspection, major defects can be eliminated on the hot-end. But hot-end sensors for bottle and can inspection should not be used only for hot-end inspection. As with any inspection machine, hot or cold, no sensor can effectively inspect all defects, and the same is true for hot-end sensors. And since every out-of-spec bottle or can produced already wastes production time and energy (and generates CO2), the focus and advantage of hot-end sensors is on defect prevention, not just automatic inspection of defective products.
The main purpose of bottle inspection with hot-end sensors is to eliminate critical defects and gather information and data. Furthermore, individual bottles can be inspected according to customer requirements, giving a good overview of the performance data of the unit, each gob or the ranker. Elimination of major defects, including hot-end pouring and sticking, ensures that products pass through hot-end spray and cold-end inspection equipment. Cavity performance data for each unit and for each gob or runner can be used for effective root cause analysis (learning, prevention) and quick remedial action when problems arise. Rapid remedial action by the hot end based on real-time information can directly improve production efficiency, which is the basis for a stable molding process.
2. Reduce interference factors
It is well known that many interfering factors (cullet quality, viscosity, temperature, glass homogeneity, ambient temperature, deterioration and wear of coating materials, even oiling, production changes, stop/start units or bottle design) affect glass manufacturing craft. These interference factors are the root cause of process variation. And the more interference factors the molding process is subjected to, the more defects are generated. This suggests that reducing the level and frequency of interfering factors will go a long way towards achieving the goal of producing lighter, stronger, defect-free and higher-speed products.
For example, the hot end generally places a lot of emphasis on oiling. Indeed, oiling is one of the main distractions in the glass bottle forming process.
There are several different ways to reduce the disturbance of the process by oiling:
A. Manual oiling: Create SOP standard process, strictly monitor the effect of each oiling cycle to improve oiling;
B. Use automatic lubrication system instead of manual oiling: Compared with manual oiling, automatic oiling can ensure the consistency of oiling frequency and oiling effect.
C. Minimize oiling by using an automatic lubrication system: while reducing the frequency of oiling, ensure the consistency of the oiling effect.
The reduction degree of process interference due to oiling is in the order of a<b<c, and c has the lowest interference to the process. The BlankRobot developed by XPAR belongs to the category c.
3. Treatment causes the source of process fluctuations to make the glass wall thickness distribution more uniform
Now, in order to cope with the fluctuations in the glass forming process caused by the above disturbances, many glass manufacturers use more glass liquid to make bottles. In order to meet the specifications of customers with a wall thickness of 1mm and achieve reasonable production efficiency, the wall thickness design specifications range from 1.8mm (small mouth pressure blowing process) to even more than 2.5mm (blowing and blowing process).
The purpose of this increased wall thickness is to avoid defective bottles. In the early days, when the glass industry could not calculate the strength of the glass, this increased wall thickness compensated for excessive process variation (or low levels of molding process control) and was easily compromised by glass container manufacturers and their customers accept.
But as a result of this, each bottle has a very different wall thickness. Through the infrared sensor monitoring system on the hot end, we can clearly see that changes in the molding process can lead to changes in the thickness of the bottle wall (change in glass distribution). As shown in the figure below, this glass distribution is basically divided into the following two cases: the longitudinal distribution of the glass and the lateral distribution.From the analysis of the numerous bottles produced, it can be seen that the glass distribution is constantly changing, both vertically and horizontally. In order to reduce the weight of the bottle and prevent defects, we should reduce or avoid these fluctuations. Controlling the distribution of the molten glass is the key to producing lighter and stronger bottles and cans at higher speeds, with fewer defects or even close to zero. Controlling the distribution of glass requires continuous monitoring of bottle and can production and measuring the operator’s process based on changes in glass distribution.
4. Collect and analyze data: create AI intelligence
Using more and more sensors will collect more and more data. Intelligently combining and analyzing this data provides more and better information to manage process changes more effectively.
The ultimate goal: to create a large database of data available in the glass forming process, allowing the system to classify and merge the data and create the most efficient closed-loop calculations. Therefore, we need to be more down-to-earth and start from actual data. For example, we know that the charge data or temperature data is related to the bottle data, once we know this relationship, we can control the charge and temperature in such a way that we produce bottles with less shift in the distribution of the glass, so that Defects are reduced. Also, some cold-end data (such as bubbles, cracks, etc.) can also clearly indicate process changes. Using this data can help reduce process variance even if it is not noticed at the hot end.
Therefore, after the database records these process data, the AI intelligent system can automatically provide relevant remedial measures when the hot-end sensor system detects defects or finds that the quality data exceeds the set alarm value. 5. Create sensor-based SOP or form molding process automation
Once the sensor is used, we should organize various production measures around the information provided by the sensor. More and more real production phenomena can be seen by sensors, and the information transmitted is highly reductive and consistent. This is very important for production!
Sensors continuously monitor the status of the gob (weight, temperature, shape), charge (speed, length, arrival time, position), temperature (preg, die, punch/core, die) to monitor the quality of the bottle . Any variation in product quality has a reason. Once the cause is known, standard operating procedures can be established and applied. Applying SOP makes the production of the factory easier. We know from customer feedback that they do feel it is getting easier to recruit new employees on the hot end because of the sensors and SOPs.
Ideally, automation should be applied as much as possible, especially when there are more and more machine sets (such as 12 sets of 4-drop machines where the operator cannot control 48 cavities well). In this case, the sensor observes, analyzes the data and makes necessary adjustments by feeding back the data to the rank-and-train timing system. Because the feedback operates on its own through the computer, it can be adjusted in milliseconds, something even the best operators/experts will never be able to do. Over the past five years, a closed loop (hot end) automatic control has been available to control gob weight, bottle spacing on the conveyor, mold temperature, core punch stroke and longitudinal distribution of glass. It is foreseeable that more control loops will be available in the near future. Based on current experience, using different control loops can basically produce the same positive effects, such as reduced process fluctuations, less variation in glass distribution and fewer defects in glass bottles and jars.
To achieve the desire for lighter, stronger, (nearly) defect-free, higher-speed, and higher-yield production, we present some ways to achieve it in this article. As a member of the glass container industry, we follow the megatrend of reducing plastic and environmental pollution, and follow the clear requirements of major wineries and other glass packaging users to significantly reduce the carbon footprint of the packaging materials industry. And for every glass manufacturer, producing lighter, stronger, (nearly) defect-free glass bottles, and at higher machine speeds, can lead to a greater return on investment while reducing carbon emissions.
Post time: Apr-19-2022